Alex Ishchenko
Machine Learning Engineer @BayOne Solutions
Signup · Get unlimited contacts
WORK HISTORY
Machine Learning Engineer @BayOne Solutions
Calgary, AB, CA
I am developing an enterprise-scale intelligent search and GPT agent platform for a leading company in the travel sector. My primary responsibility is to architect and implement large-scale data pipelines for processing and indexing vast amounts of structured and unstructured data. The goal is to achieve high-precision hybrid search by leveraging techniques such as data enrichment, semantic reranking, query expansion, embedding optimization, vector quantization, metadata-driven filtering, and domain-specific relevance tuning.Responsibilities- Building comprehensive data governance frameworks utilizing PySpark and medallion architecture (Bronze, Silver, Gold layers) to ensure data quality, lineage, and compliance across the organization- Developing robust document extraction pipelines for PDF, DOCX, and PPT files based on Azure Document Intelligence- Developing offline-batch and online RAG solutions based on Azure AI and ElasticSearch- Training content classification models by using Azure AI classification models, Sentence Transformers models- Working on approaches for chunking content, embeddings generation by using QWEN embeddings models- Improving search relevance results by combining variety of reranking approaches and models- Building entire company level evaluation framework and utilizing LLM-as-a-judge technique- Using variety of inference engines for deployment vLLM, TGI, TEI- Using CLIP models for multi modal data- Supervised fine-tuning utilizing PEFT(LoRA) technique and using Unsloth, TRL library for the PHI-¾ models- Building 3-tier caching system utilizing Redis
EDUCATION
Simon Kuznets Kharkiv National University of Economics
Master's degree, Computer Science
ABOUT ALEX ISHCHENKO
I am a Machine Learning Engineer and Data Engineer with a strong background in software engineering, MLOps, and data science. Currently, I work for a large travel company where I specialize in designing, training, and deploying large-scale AI systems. I build robust data governance and analytics pipelines using PySpark, Databricks, Delta Lake, Apache Iceberg, implementing partitioning strategies,data skew mitigation to ensure scalable and efficient data processing.My current focus is on LLMs, LMMs, where I develop end-to-end RAG, MRAG,hybrid enterprise search, recommendation systems using Azure AI Search and Elasticsearch. I have practical knowledge of LLM inference engines including TEI, TGI, and vLLM, and specialize in fine-tuning techniques such as supervised fine-tuning using PEFT (LoRA/QLoRA) with frameworks like TRL,ms-swift,LLaMA Factory. My expertise extends to model quantization, dataset preparation for supervised fine-tuning, and implementing production monitoring systems with data quality and drift detection. Solid experience in time series forecasting.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.